Bug 1447571

+++ This bug was initially created as a clone of Bug #1442950 +++

Description of problem:

Any application using glfs handles (like NFS-Ganesha) may need to do lookup on each dirent returned by readdir/readdirp operation to create handles. But since the lookup in glusterfs is very costly operation (as need to be sent to all replica bricks), as the directory gets larger or the replica count increases, the time taken for single readdir operation to complete can take hours of time.

To avoid that we need an extended readdirp API which can return handles along with dirent stat as part of its reply.

Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

--- Additional comment from Worker Ant on 2017-04-18 04:54:17 EDT ---

REVIEW: https://review.gluster.org/15663 (WIP/gfapi/handleops: Introducing READDIRPLUS_EXT() fop for handleops) posted (#6) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-04-21 08:42:06 EDT ---

REVIEW: https://review.gluster.org/15663 (gfapi/handleops: Introducing glfs_h_xreaddirplus_r() fop for handleops) posted (#7) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-04-21 08:45:48 EDT ---

REVIEW: https://review.gluster.org/15663 (gfapi/handleops: Introducing glfs_h_xreaddirplus_r() fop for handleops) posted (#8) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-04-26 03:44:21 EDT ---

REVIEW: https://review.gluster.org/15663 (gfapi/handleops: Introducing glfs_h_xreaddirplus_r() fop for handleops) posted (#9) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-04-27 05:08:30 EDT ---

REVIEW: https://review.gluster.org/15663 (gfapi/handleops: Introducing glfs_h_xreaddirplus_r() fop for handleops) posted (#10) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-04-27 05:43:45 EDT ---

REVIEW: https://review.gluster.org/15663 (gfapi/handleops: Introducing glfs_h_xreaddirplus_r() fop for handleops) posted (#11) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-04-28 07:55:10 EDT ---

REVIEW: https://review.gluster.org/15663 (gfapi/handleops: Introducing glfs_h_xreaddirplus_r() fop for handleops) posted (#12) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-05-01 22:55:09 EDT ---

REVIEW: https://review.gluster.org/15663 (gfapi/handleops: Introducing glfs_xreaddirplus_r() fop for handleops) posted (#13) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-05-01 23:09:29 EDT ---

REVIEW: https://review.gluster.org/15663 (gfapi/handleops: Introducing glfs_xreaddirplus_r() fop for handleops) posted (#14) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-05-02 01:49:10 EDT ---

REVIEW: https://review.gluster.org/15663 (gfapi/handleops: Introducing glfs_xreaddirplus_r() fop for handleops) posted (#15) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-05-02 11:54:05 EDT ---

REVIEW: https://review.gluster.org/15663 (gfapi/handleops: Introducing glfs_xreaddirplus_r() fop for handleops) posted (#16) for review on master by soumya k (skoduri)

--- Additional comment from Worker Ant on 2017-05-02 19:33:16 EDT ---

COMMIT: https://review.gluster.org/15663 committed in master by Jeff Darcy (jeff.us) 
------
commit 41000cd0b57a81c4ace4a1d3da0fcc352a11f146
Author: Soumya Koduri <skoduri>
Date:   Fri Apr 21 16:30:20 2017 +0530

    gfapi/handleops: Introducing glfs_xreaddirplus_r() fop for handleops
    
    Its known that readdirplus operation fetches stat as well for each of the
    dirents. But often applications may need extra information, like for eg.,
    NFS-Ganesha which operates on handles needs handles for each of those
    dirents returned. So this would require extra calls to the backend, in this
    case LOOKUP (which is very expensive operation) resulting in very low
    readdir performance.
    
    To address that introducing this new API using which applications can
    make request for any extra information to be returned as part of
    readdirplus response.
    
    Currently this new api returns stat and handles as demanded by application.
    The synopsis of the API is noted in glfs.h.
    
    @todo:
    * Enhance test script using this new API
    
    Below were the perf results on single brick volume with and without
    these changes -
    
    Dataset used -
    10*100 directories and each directory containing 100 empty files.
    
    I used NFS-Ganesha application to test these changes -
    >for i in {1..5}; do systemctl restart nfs-ganesha; sleep 10; mount -t nfs -o vers=4 localhost:/brick_vol /mnt; cd /mnt; echo "ITERATION$i"; date; find . > tmp-nfs.log; date; cd /; umount /mnt; sleep 2; done;
    
    Without these changes -
    ITERATION1
    Mon Mar 20 17:22:26 IST 2017
    Mon Mar 20 17:23:18 IST 2017
    ITERATION2
    Mon Mar 20 17:23:39 IST 2017
    Mon Mar 20 17:24:28 IST 2017
    ITERATION3
    Mon Mar 20 17:24:49 IST 2017
    Mon Mar 20 17:25:36 IST 2017
    ITERATION4
    Mon Mar 20 17:30:57 IST 2017
    Mon Mar 20 17:31:37 IST 2017
    ITERATION5
    Mon Mar 20 17:31:57 IST 2017
    Mon Mar 20 17:32:40 IST 2017
    [root@dhcp35-197 /]#
    
    On an average ~46.2 sec
    
    With these changes applied -
    ITERATION1
    Mon Mar 20 17:35:03 IST 2017
    Mon Mar 20 17:35:15 IST 2017
    ITERATION2
    Mon Mar 20 17:35:36 IST 2017
    Mon Mar 20 17:35:46 IST 2017
    ITERATION3
    Mon Mar 20 17:36:06 IST 2017
    Mon Mar 20 17:36:17 IST 2017
    ITERATION4
    Mon Mar 20 17:41:38 IST 2017
    Mon Mar 20 17:41:49 IST 2017
    ITERATION5
    Mon Mar 20 17:42:10 IST 2017
    Mon Mar 20 17:42:20 IST 2017
    
    On an average ~10.8 sec
    
    Updates #174
    BUG: 1442950
    Change-Id: I0f74f74dc62085ca4c4a23c38e3edc84bd850876
    Signed-off-by: Soumya Koduri <skoduri>
    Reviewed-on: https://review.gluster.org/15663
    Smoke: Gluster Build System <jenkins.org>
    NetBSD-regression: NetBSD Build System <jenkins.org>
    Reviewed-by: Niels de Vos <ndevos>
    CentOS-regression: Gluster Build System <jenkins.org>